자료유형  

 학위논문

Control Number  

0016932101

International Standard Book Number  

9798379605018

Dewey Decimal Classification Number  

610

Main Entry-Personal Name  

Talkar, Tanya.

Publication, Distribution, etc. (Imprint  

[S.l.] : Harvard University., 2023

Publication, Distribution, etc. (Imprint  

Ann Arbor : ProQuest Dissertations & Theses, 2023

Physical Description  

1 online resource(204 p.)

General Note  

Source: Dissertations Abstracts International, Volume: 84-12, Section: B.

General Note  

Advisor: Quatieri, Thomas.

Dissertation Note  

Thesis (Ph.D.)--Harvard University, 2023.

Restrictions on Access Note  

This item must not be sold to any third party vendors.

Summary, Etc.  

요약Neurological disorders such as Autism Spectrum Disorder (ASD) and Parkinson's Disease (PD) are typically associated with observed motor difficulties in speech production and in fine-motor tasks (including oculo-motor tasks), each condition with its own characteristic motor challenges. Clinical assessments of these motor impairments, however, can be subjective and miss subtle and specific abnormalities. Given the use of these assessments in determining therapeutic trajectories, there is a need to develop objective measures for clinicians to assess and understand speech- and fine-motor impairments in ASD and PD. In addition, for personalized tracking of progress due to therapies, simple, at-home assessments can inform clinicians if there are changes that occur between check-up appointments. To address this, our overarching goal is to derive objective measures of motor coordination, to be applied across motor modalities, that can be used to support clinical decision making. To address this, we have employed a physiologically-based analysis framework to act as a proxy for assessing motor coordination. We develop an off-body data collection system that administers protocols involving speech, hand drawing, and eyetracking tasks. From recordings on this platform, low-level signals that represent motor modalities are used in constructing correlation matrices that aid in determining the complexity of the signals. To provide interpretability, we simulate specific signal changes which can lead to higher or lower complexity in the signals. We describe how the correlation-based analysis can extend to correlations within and across speech production subsystems (articulatory, laryngeal, and respiratory), and in finemotor modalities. The methodology is then applied to four sub-studies. We focus on ASD, a neurodevelopmental disorder, and PD, a neurodegenerative disorder, at differing levels of severity, to develop an understanding of the commonalities and differences of how speech- and fine-motor deficits present across these disorders. First, we analyze speech, handwriting, and eye gaze patterns in a pilot study with minimally-verbal adults with ASD. This novel study highlights the heterogeneity of the minimally-verbal population and suggests features that may lead to the differences we see between minimally-verbal individuals and neurotypical controls. In line with our hypotheses, we see lower complexity of articulatory signals and shape drawing trajectories in the minimally-verbal ASD group as compared to neurotypical controls. However, interestingly, we witness higher complexity of eye gaze trajectories and handwriting trajectories, highlighting task- and modality- related differences in motor coordination in this population. We additionally apply our methodology to a study conducted with at-home, naturalistic recordings of minimally-verbal children to aid in understanding differences in speech motor coordination between different communication intents such as request or frustration. Our perceptual assessments of the speech production subsystems involved in differentiating between self-talk, request, and delighted utterances as compared to frustrated and dysregulated utterances are quantified in the correlation-based features, highlighting difference primarily in the laryngeal and articulatory subsystems. With highly-verbal children with ASD, we discuss subtle differences in speech, handwriting, and eye gaze patterns that aid in discriminating between this population and neurotypical controls, using our correlation-based methodology. Correlation-based features are able to discriminate between ASD and neurotypical groups with AUCs 0.80, and highlight lower complexity of coordination in children with ASD. Features derived from handwriting and maze tracing tasks discriminate between ASD and neurotypical groups with AUCs 0.86 and 0.91, respectively, and show high complexity of underlying drawing trajectories in the ASD group as compared to the neurotypical group. Finally, features derived from free speech and sustained vowel tasks aid in prediction of expressive vocabulary scores, suggesting a link between speech motor ability and language ability. Finally, in individuals with PD, we utilize a single- and dual-task paradigm, where speech and hand drawing tasks are completed in isolation or simultaneously, to aid in detection of mild cognitive impairment (MCI). Contrary to our initial hypothesis, the use of a cognitively difficult dual-task paradigm does not lead to improved discrimination between an MCI group and a group with no cognitive impairment (NCI) as compared to a single-task speech-based paradigm. Specifically, there are a large number of correlation-based features from a reading-based single-task paradigm that lead to AUCs of ≥ 0.70 in discriminating between the two groups. The correlation-based framework developed in this work, as well as the applications in which it has been utilized, provide a foundation for use of objective measures of motor coordination in clinical-based decision-making and for off-body remote monitoring of motor ability in individuals with a range of conditions. This work additionally presents protocols and data collection systems which aid in off-body assessment of motor coordination. Future work will entail diving deeper into feature interpretability for clinical use, improved data collection platforms for at-home use, and improved understanding of the relationships across motor modalities.

Subject Added Entry-Topical Term  

Biomedical engineering.

Subject Added Entry-Topical Term  

Neurosciences.

Subject Added Entry-Topical Term  

Speech therapy.

Index Term-Uncontrolled  

Machine learning

Index Term-Uncontrolled  

Motor coordination

Index Term-Uncontrolled  

Neurodevelopmental disorders

Index Term-Uncontrolled  

Neurological disorders

Index Term-Uncontrolled  

Signal processing

Index Term-Uncontrolled  

Speech production

Added Entry-Corporate Name  

Harvard University Medical Sciences

Host Item Entry  

Dissertations Abstracts International. 84-12B.

Host Item Entry  

Dissertation Abstract International

Electronic Location and Access  

로그인을 한후 보실 수 있는 자료입니다.

Control Number  

joongbu:640851